It’s been hard to get a decent night’s sleep lately.

Stories of the Government’s mismanagement of its quarantine facilities and of the pandemic worsening abroad make for restless nights thinking through safer ways of running those operations. Or, on luckier nights, pleasant dreams of Peter Capaldi’s fictional government fixer from The Thick of It, Malcolm Tucker, arriving at the Ministry of Health to berate them all into submission with exactly the right Glaswegian curses.

But economists can have other nightmares too.

I started worrying about this one at the Budget lockup, when Finance Minister Robertson steadfastly refused to say whether the Government’s position about prudent debt levels had changed, or whether he expected that the Government’s path back to more conventionally prudent levels would just take an awfully long time.

So welcome to my nightmare.

In 2023, New Zealand’s net debt to GDP ratio is projected to hit about 54 percent. During normal times, Treasury has seen a 60 percent debt-to-GDP ratio as being about the outer limit of what the country can handle. That’s why both parties maintained a substantial buffer in case of disaster.

In my nightmare, the Alpine Fault opens up in 2023, but as a foreshock. The earthquake experts at GNS Science, on studying it, expect a second, bigger earthquake to come on the Fault in 2024, and to trigger the Wellington Fault at the same time. In 2024, the Big One comes. The Alpine Fault triggers different building failures in Christchurch than it experienced in 2011, and Wellington is close to a write-off.

That is nightmare enough on its own. While the Government has had its inquiry into the failings of the governmental response in Christchurch, too few of the lessons of Christchurch have been applied. The terrible natural disaster is likely to be compounded by infuriatingly avoidable policy failures. And any earthquake response in Wellington will be far more difficult than it was in Christchurch because of the two cities’ differences in topography. It is easy to get into and out of Christchurch. That will not be true in Wellington.

But it does get worse. The Christchurch earthquakes came at a time when the then-government was easily able to borrow more to deal with the crisis. It could have borrowed more than it did, and arguably should have.

Come the quake in 2023 and 2024, the conversations with international lenders could be rather more difficult. In the same way that it is much easier to get a substantial top up on your mortgage if you’ve only borrowed 20 percent of the value of your house than if you’re already fully leveraged, it will be harder for the Government to take on the debt it needs to deal with the earthquakes.

In my nightmare scenario, Treasury was rather optimistic in thinking borrowing 20 percent of GDP would be enough to see us through a big Wellington earthquake.

Homeowners are heavily insured, but the first $150,000 in damage is covered by the Earthquake Commission. The Earthquake Commission currently holds $6.2 billion in reinsurance coverage, with the Crown liable for any amount that the EQC cannot cover.

There are about 200,000 dwellings in the Greater Wellington area, and about 150,000 more in Christchurch. Even if each, on average, only costs EQC $100,000, that’s still about $35 billion in potential EQC backstop liabilities over and above what it has in reinsurance coverage – over 10 percent of GDP all on its own. And EQC’s land coverage will be added in. And all of the other costs that come with a natural disaster, from business and family support through to infrastructure and relocating government agencies.

While international credit agencies and international debt markets showed little concern in 2020 about the debt that the Government took on to deal with Covid, those measures focused on the Government’s ability to take on and service that debt. Not the additional substantial debt the Government might need to take on in a new crisis.

After the 2023 foreshock, the Government will put a good face on it, reminding everyone of New Zealand’s strong reputation for fiscal prudence and that the country is seen as a very sound borrower.

But quietly, in the background, the usual larger buyers of government debt may be telling the Government, a bit sheepishly, that they’re just not able to help us out if there is a bigger earthquake in 2024. The risk is too great if the Government needs to access another 25 percent or 30 percent of GDP on top of its current borrowing.

Come the bigger earthquake in 2024, the Government cannot find buyers for its debt on normal terms. Its options are not good. JP Morgan, or another lender, comes round and makes the Government an offer. New Zealand’s borrowing will be expensive, and will be denominated in US Dollars, Euros and Yen because everyone sees that the pressures to inflate or devalue will be strong. A country that is heavily in debt in its own currency is one that can be tempted to print a lot of money to reduce the value of that debt. The interest rates will be higher than New Zealand has been used to as well. And similar conditions will apply to older debt that the Government needs to roll over.

In the 1980s, financing government debt cost over 6 percent of GDP, and required rather strict measures to get things under control. Dealing with a substantial natural disaster, on top of current debt levels, could push New Zealand back into that kind of a world.

In the cold light of morning, I remind myself that the nightmare scenario is not all that probable. GNS forecasts only about a 0.83 percent chance that the Wellington fault opens up in any year – but it could happen tomorrow too.

And I remind myself that, surely, the Government’s Debt Management Office will have been exploring these kinds of scenarios with international lenders – if they’d be nervous about extending further credit in that kind of nightmare scenario, the Debt Management Office would surely be advising the Government of those risks.

And surely the Debt Management Office has been looking at the case for issuing some of its new pandemic debt as Catastrophe Bonds. Catastrophe Bonds pay investors a higher interest rate in exchange for the cancelling of the bond if a specified disaster strikes. Paying a bit more in interest on debt now as insurance against that nightmare scenario could easily be worth it. It would provide headroom in the case in which it would most be needed.

But if debt-to-GDP ratios are only down to 42 percent by 2034 and none of that new debt is issued as catastrophe bonds, that’s fourteen years of those 0.83 percent annual risks compounding – and with 2034 debt levels that still leave too little headroom for dealing with any substantial additional crisis. Or, at least, too little headroom relative to what has been the bipartisan consensus over decades.

Taking on debt in a crisis absolutely makes sense. Spending more than is necessary, though, builds in immense fragility. A pandemic does not make other crises less likely.

Pleasant dreams.